Antennas · Volume 8
Vertical Monopoles (Fixed)
Quarter-wave verticals, ground planes, full-size HF verticals (Hustler 4/5/6-BTV, R-7000, GAP), shortened loaded verticals, the half-square, the radial field as a load-bearing element
Contents
1. About this volume
A vertical monopole is, in a sense, the simplest possible antenna: a single λ/4 conductor standing upright, fed against a ground reference at its base. There is no second conductor to terminate, no balanced feed, no symmetry to maintain. And yet vertical antennas are arguably the most misunderstood family in amateur radio — because their performance lives and dies by the radial field, an invisible structure below the antenna that does at least as much radiating work as the visible whip does.
This volume is about fixed verticals — antennas planted at a specific location with a permanent radial system, where you have the time and space to do the ground system right. Portable and mobile verticals (mag-mounts, ham-sticks, MP-1, J-pole, Slim Jim, rubber ducks) are a structurally different problem with a different set of tradeoffs and live in Vol 9. The dividing line is whether the operator can deploy a proper radial system: if yes, this volume’s antennas dominate; if no, Vol 9’s compromises take over.
The fixed-vertical’s signature advantage is low-angle radiation. A horizontal dipole at residential height has its peak elevation lobe at 30–60°, which is fine for NVIS and short-skip but suboptimal for DX paths that come in at 5–20° elevation. A quarter-wave vertical over a competent radial field puts its peak at 20–30° elevation and has substantial radiation down to 5° — the angles where transcontinental and intercontinental HF propagation actually arrives. This single advantage justifies the vertical’s existence even when the radial system is a hassle. The catch — and it’s a big one — is that a vertical without a competent ground system isn’t a low-angle DX antenna; it’s a high-angle inefficient noise-pickup antenna. The radial field is the difference between a hero and a disappointment.
2. Geometry & theory — the quarter-wave monopole over ground
2.1 The image-plane derivation
A quarter-wave monopole over a perfect ground is electrically equivalent to a half-wave dipole rotated 90° and split — with one half buried below ground, mirrored by the ground’s electromagnetic image. The math is the method-of-images result from electromagnetism: a perfectly conductive ground plane reflects the antenna’s image as if a second, mirror-symmetric antenna existed below. The above-ground monopole + its image = a complete half-wave dipole.
╱ monopole, λ/4 tall
╱
● feedpoint (fed against ground)
═══ ═══ ═══ ═══ ground plane (radial system)
● image feedpoint (virtual)
╲
╲ image antenna (virtual, λ/4 long)
What this tells you immediately:
- The feedpoint impedance is half the dipole’s: 73/2 ≈ 36 Ω.
- The gain is the same in the upper hemisphere as a dipole’s half-pattern: the monopole sees only the upper half-space, so its dBi is 3 dB higher in that hemisphere than the equivalent dipole. Stated as 5.15 dBi in free space (or 2.15 dBi + 3 dB from being one-sided), which is the canonical quarter-wave monopole gain figure.
- The pattern is omnidirectional in azimuth (no figure-8) and donut-shaped in 3D, with the donut’s hole pointing straight up. The vertical is a 360°-azimuth antenna by construction.
- The polarization is vertical.
2.2 The catch: real ground is not perfect
Real ground is lossy. The image-plane derivation assumes a perfect conductor; real soil has finite conductivity (10⁻³ S/m for dry sand, 10⁻¹ S/m for wet farmland, 5 S/m for salt water) and finite dielectric constant. Real-ground currents flowing back to the feedpoint dissipate as heat — this is the “ground loss” that eats your transmitted power.
A radial field’s purpose is to replace the lossy real ground with a low-loss artificial conductor for the near-field return-current zone. The far-field radiation pattern is established by the upper-hemisphere geometry (whip + radial system); the near-field return currents are what the radial field intercepts and routes back to the feedpoint with minimal loss.
The implication: without a competent radial field, a quarter-wave vertical loses 3–10 dB of efficiency to ground heating. With 60 buried radials over typical soil, the loss drops to 0.5–1 dB. With 4 elevated radials (counterpoise wires lifted off ground at λ/4 height), the loss drops to <0.3 dB — better than the buried-radial case because the elevated radials don’t see lossy ground currents at all.
2.3 The standing-wave picture on a vertical
Like the half-wave dipole, the quarter-wave vertical is a resonator. The current distribution is a quarter-cycle sinusoid: maximum at the feedpoint (base), zero at the top. The voltage distribution is its 90°-shifted partner: zero at the feedpoint, maximum at the top.
top: V high, I = 0
●
│ voltage current
│ ────► ◄────
│
│
│
│
● feedpoint: I high, V low
│
│ Z = 36 Ω over perfect ground
│
═══════ ground plane (radial system)
radials extend λ/4 from base, all in phase
Two consequences flow directly from this picture:
- The top is at high voltage — needs insulation from any nearby objects. Many shortened verticals have a capacitive top-hat at the high-voltage point, which works with the voltage maximum rather than fighting it.
- The feedpoint is at high current — needs low-resistance connection. The radial connection to the feedpoint should be a low-impedance bond, multiple radials soldered or clamped to a common return point, with a clean RF path to the coax shield.
2.4 Quarter-wave vertical length table
Using the standard L_feet = 234 / f_MHz formula (half of the dipole formula, because monopole = half dipole):
| Band | Design freq | λ/4 free-space | Cut at k=0.95 (#10 wire) | Cut at k=0.93 (1/2″ tubing) |
|---|---|---|---|---|
| 160 m | 1.900 MHz | 39.47 m (129.5 ft) | 37.50 m (123.1 ft) | 36.71 m (120.5 ft) |
| 80 m | 3.650 MHz | 20.55 m (67.4 ft) | 19.52 m (64.0 ft) | 19.11 m (62.7 ft) |
| 60 m | 5.358 MHz | 13.99 m (45.9 ft) | 13.29 m (43.6 ft) | 13.01 m (42.7 ft) |
| 40 m | 7.150 MHz | 10.48 m (34.4 ft) | 9.96 m (32.7 ft) | 9.75 m (32.0 ft) |
| 30 m | 10.125 MHz | 7.40 m (24.3 ft) | 7.03 m (23.1 ft) | 6.88 m (22.6 ft) |
| 20 m | 14.175 MHz | 5.29 m (17.4 ft) | 5.03 m (16.5 ft) | 4.92 m (16.1 ft) |
| 17 m | 18.118 MHz | 4.14 m (13.6 ft) | 3.93 m (12.9 ft) | 3.85 m (12.6 ft) |
| 15 m | 21.225 MHz | 3.53 m (11.6 ft) | 3.36 m (11.0 ft) | 3.28 m (10.8 ft) |
| 12 m | 24.940 MHz | 3.01 m (9.86 ft) | 2.86 m (9.37 ft) | 2.80 m (9.18 ft) |
| 10 m | 28.500 MHz | 2.63 m (8.63 ft) | 2.50 m (8.20 ft) | 2.45 m (8.04 ft) |
| 6 m | 50.250 MHz | 1.49 m (4.89 ft) | 1.42 m (4.65 ft) | 1.39 m (4.55 ft) |
A 40 m vertical is 33 ft of conductor — a fiberglass push-up mast with wire wrapped along its length, or aluminum tubing, or just a tall stick of conductor against gravity. A 160 m vertical is 123 ft — towering, and almost always built as a wire vertical hung from a high support point or as a heavily-loaded short vertical (see §8).
3. Feedpoint impedance and matching
3.1 The 36 Ω feedpoint over perfect ground
A quarter-wave monopole over a perfect ground (60+ buried radials over decent soil approximates this) presents 36 + j21 Ω at exact resonance — half the dipole’s 73 + j42. Trim to physical resonance (remove the inductive reactance) and the feedpoint sits at 36 Ω resistive.
On 50 Ω coax this is SWR = 50/36 = 1.39:1 — completely usable without matching, and modern HF rigs deliver full power into 1.4:1 with no protection-circuit fold-back. The 0.12 dB of mismatch loss is inaudible. No matching network is required for a textbook quarter-wave vertical over a competent radial field on 50 Ω coax — the most common simplification trap is to assume you need a transformer when you don’t.
3.2 Real-world impedance drift
Real-ground operation and real radial fields shift the impedance:
| Radial system | Feedpoint Z (typical) | SWR on 50 Ω | Notes |
|---|---|---|---|
| Perfect ground (theoretical) | 36 Ω | 1.39:1 | Reference |
| 60+ buried radials, good soil | 36–42 Ω | 1.2–1.4:1 | The “right” radial system |
| 32 buried radials, good soil | 40–50 Ω | 1.0–1.25:1 | Adequate — typical good install |
| 16 buried radials, good soil | 50–65 Ω | 1.0–1.3:1 | Sparse — SWR misleadingly good (ground loss “matches” the resistance up) |
| 4 buried radials only | 65–90 Ω | 1.3–1.8:1 | Inadequate — SWR low because of ground loss |
| 4 elevated radials at λ/4 height | 25–45 Ω | 1.1–1.4:1 | The “no buried radials” alternative |
| Ground-mounted with NO radials | 80–150 Ω | 1.6–3:1 | Pathological — most of the power is heating soil |
This table contains an important counterintuitive truth: a vertical with too few buried radials shows a better SWR than a vertical with too many. The reason: ground loss adds a resistive component to the feedpoint impedance, lifting the 36 Ω toward 50 Ω. The antenna with 4 radials reads “1.5:1 SWR — looks great!” but most of the power is being dissipated as heat in the soil. The antenna with 60 radials reads “1.2:1 SWR — a bit low” because the radial field has successfully eliminated the ground loss and exposed the true 36 Ω feedpoint.
SWR alone is not a measure of antenna efficiency. The radial field’s quality is what matters; SWR is just one indirect symptom.
3.3 Matching options for the 36 Ω feedpoint
If you want to clean up the 1.4:1 SWR (and there’s no operational reason to do this — it’s purely aesthetic), three matching schemes work:
- Shunt-L (Beverage match): a small inductor in parallel with the feedpoint shifts the impedance up. Cheap, lossless, narrow-band.
- Series-C / shunt-L L-network: a series capacitor + shunt inductor matches 36 Ω to 50 Ω at a single design frequency. Narrow-band, dead-simple, $5 in parts.
- 1.5:1 transformer (unun): 36 Ω to 50 Ω is exactly 1.4× ratio, and a 1.5:1 transmission-line transformer (3-turn bifilar on a small toroid) lands close enough. Balun Designs 1112a-ATU is the canonical commercial part ($85); homebrew is straightforward.
- Sloped radials (elevated): lifting the radials off ground and sloping them downward at 30–45° from horizontal raises the feedpoint Z toward 50 Ω naturally. Lift the radials, get a free match, lose nothing in the bargain.
The sloped-radial trick is the most elegant: an elevated ground-plane with radials sloped 30–45° from horizontal presents 50 Ω at the feedpoint without a matching network. This is why most commercial elevated ground-plane verticals come with adjustable-angle radial brackets — the angle adjustment is the impedance match.
3.4 A 1:1 current choke at the feedpoint
Even with a properly matched vertical, the coax shield is at the same potential as the antenna’s ground reference (the radial system common point). Common-mode currents can flow on the coax shield exterior — though they’re less common with verticals than with dipoles (verticals are naturally “fed against ground,” so the asymmetry that drives common-mode current on a dipole is partly inherent here).
A 1:1 current choke at the feedpoint (or just below, where the coax exits the radial connection) suppresses common-mode current. The same Mix-31 toroid or string-of-beads design used for dipole BALUNs works here. The choke is more important for elevated verticals than for ground-mounted verticals (elevated verticals’ radial counterpoise is more sensitive to common-mode pickup).
4. Radiation pattern — the low-angle vertical advantage
4.1 The vertical’s pattern in 3D
A quarter-wave vertical’s free-space pattern is the upper half of a half-wave dipole’s torus — the same donut shape, but with the lower half cut off by the ground plane. The pattern is:
- Omnidirectional in azimuth — uniform around the 360° horizon.
- Toroidal in 3D — donut with the hole pointing straight up (zenith null).
- Peak elevation ~25–30° above horizon over ideal ground.
- Peak gain ~5.15 dBi in free-space terms, or ~3 dBi over typical real ground.
Vertical's elevation pattern (cross-section, side view):
↑ zenith (null — 90° elevation)
╱─╲ ╱─╲
╱ ╲ ╲
╱ ╲ ╲
╱ peak ╲
╱ elevation ╲
╱ 25-30° ╲
●═══════════════════════● horizon (0°)
feedpoint at base
═══════════════════════ ground plane
Compared to a horizontal dipole at h = 0.25λ (which peaks at 50° elevation and has substantial radiation only above 30°), the vertical’s peak at 25° puts the signal where DX propagation paths actually arrive. This is the low-angle advantage.
4.2 Elevation pattern by frequency / soil quality
The vertical’s peak elevation depends on ground quality:
| Soil | Conductivity (S/m) | Dielectric ε | Peak elevation | Peak gain |
|---|---|---|---|---|
| Salt water | 5.0 | 80 | 18° | 5.5 dBi |
| Wet farmland | 0.03 | 25 | 22° | 4.5 dBi |
| Typical soil | 0.005 | 13 | 28° | 3.5 dBi |
| Dry rocky soil | 0.001 | 8 | 38° | 1.5 dBi |
| Dry sand | 0.0001 | 4 | 45° | -1.0 dBi |
| Desert (very dry) | 0.00005 | 3 | 50° | -2.0 dBi |
The 6.5 dB swing in peak gain across the soil-quality range is real and is the primary reason vertical-vs-dipole “which is better” debates never converge — a vertical at a salt-water beach destroys a dipole at the same location; the same vertical in a Nevada desert is a disaster. The radial system can only partially compensate for poor soil — radials route near-field current efficiently, but far-field reflection is still soil-determined, and soft soil with poor far-field reflection reduces the low-angle peak by a few dB regardless of how many radials you bury.
4.3 Azimuth — uniform, but ground-feature dependent
A vertical’s azimuth pattern is theoretically uniform 360°. Real installations introduce asymmetries: a metal-roof building 20 m east of the vertical creates a passive reflector that adds gain east (and reduces it west); a tower 30 m north creates a similar parasitic effect. A “vertical in the middle of nothing” approximates the textbook omnidirectional pattern; a “vertical in a residential lot” usually doesn’t.
The implication: when siting a vertical, leave at least 1 wavelength of clear space around it on the most-important-direction azimuth. For a 40 m vertical that’s 40 m of clear space — often impractical in residential lots, but knowing the constraint lets you make informed compromises (put the vertical on the side of the lot closest to the most-important DX direction).
4.4 The vertical noise problem
Verticals are receive-noisier than horizontal dipoles, in two distinct ways:
- Vertically-polarized local interference dominates: power-line discharge, switching power supplies, plasma TVs, switching LED drivers — most local-electrical noise is vertically polarized, and a vertical antenna picks it up preferentially. A horizontal dipole rejects vertical-pol noise by ~6–10 dB.
- Low-angle reception catches both signal and noise: the same low-angle pattern that helps with DX also picks up low-angle noise sources (distant electrical noise, atmospheric statics on the lower HF bands).
The receive noise penalty is real and usually 6–15 dB on the HF bands. Many serious DX stations use a vertical for transmit and a separate Beverage or low-noise loop antenna (Vol 15) for receive — exactly because of this asymmetry.
5. Frequency response & SWR curve
5.1 Single-band bandwidth
A quarter-wave vertical’s bandwidth depends on the conductor’s diameter (like a dipole — fatter = wider). For typical amateur installations:
| Vertical type | Bandwidth (2:1 SWR) | Notes |
|---|---|---|
| #14 wire vertical | 4–6% | Narrower than a dipole because the radial system loads the feedpoint |
| 1/2″ aluminum tubing | 6–8% | The standard amateur build |
| 2″ commercial vertical (Hustler R-7000) | 8–12% | Wider — better commercial build |
| Cage vertical (multiple wires) | 10–15% | Wide; rare in HF amateur use |
For 40 m, a typical 1/2″ aluminum tubing vertical has 2:1-SWR bandwidth of ~400 kHz — covers the whole band easily. For 80 m, bandwidth becomes a real constraint: a #14 wire 80 m vertical covers only ~150 kHz at 2:1 (similar to the dipole problem in Vol 6 §5.1). Multi-band trap verticals (§7) split this further, with each band’s bandwidth narrowing to ~3% due to trap loading.
5.2 Multi-band trap vertical SWR profile
A typical Hustler 6-BTV (10/12/15/17/20/30/40 m) shows the following SWR profile at the feedpoint, with adequate radial system:
| Band | SWR minimum | 2:1 BW | Tuner needed? |
|---|---|---|---|
| 80 m | (not covered) | — | — |
| 40 m | 1.5–2.0:1 at 7.15 MHz | 100 kHz | Marginal (depends on rig) |
| 30 m | 1.5–2.5:1 at 10.12 MHz | 70 kHz | Marginal |
| 20 m | 1.3–1.8:1 at 14.15 MHz | 150 kHz | No |
| 17 m | 1.4–2.0:1 at 18.10 MHz | 100 kHz | Borderline |
| 15 m | 1.5–2.2:1 at 21.20 MHz | 120 kHz | Marginal |
| 12 m | 1.6–2.5:1 at 24.94 MHz | 80 kHz | Borderline |
| 10 m | 1.3–1.8:1 at 28.50 MHz | 250 kHz | No |
The 40 m bandwidth is narrowest (3% — typical trap-loaded behavior). Operators who want 80 m coverage need either an add-on resonator coil (Hustler RM-80) or a separate 80 m antenna.
6. Ground plane vs elevated radials — efficiency tradeoff
This is the decision for a fixed vertical installation, and it deserves a section of its own.
6.1 Buried radial fields
The textbook answer for a ground-mounted quarter-wave vertical is 120 buried radials, each λ/4 long, laid radially outward from the feedpoint at ~3° azimuth spacing. The 120-radial field originated in commercial-broadcast engineering (BC AM stations) and represents the empirical knee where additional radials produce diminishing returns.
For amateur installations, the practical rule-of-thumb:
| # buried radials | Efficiency (typical soil) | Effort | Marginal value |
|---|---|---|---|
| 0 | 5–20% | None | — |
| 4 | 25–40% | One afternoon | +20–30 percentage points |
| 8 | 40–55% | A weekend | +15 pp |
| 16 | 55–70% | Two weekends | +15 pp |
| 32 | 70–82% | Multiple weekends | +12 pp |
| 60 | 82–90% | Serious commitment | +8 pp |
| 120 | 90–95% | Broadcast-grade | +5 pp |
| 240 | 92–96% | Diminishing returns | +1–2 pp |
The take-away: 0 → 8 radials gains you 30+ percentage points of efficiency, the biggest single improvement you can make. 8 → 32 gains another 30 points. 32 → 120 gains 15 more. Diminishing returns after that. Most serious amateur installations should target 32–60 radials for a good cost/effort/performance balance; 120+ is overkill except for low-band DX stations.
Radial length matters too: at λ/4 each, the standard advice is “as long as λ/4 if you have the space; if not, shorter radials work but at reduced efficiency.” A 32-radial field with λ/8 radials is roughly equivalent to a 16-radial field with λ/4 radials.
Buried-radial best practice (covered in depth in Vol 20 §3):
- Use #14 or #16 copper-clad steel wire (“CCS” — the broadcast-industry standard)
- Lay the radials directly on the surface (or just below — 2-3 cm deep is fine; deeper doesn’t help)
- Bond all radials to a common feedpoint ground bus
- Don’t bother insulating or weatherproofing — buried radials self-protect
6.2 Elevated radials
For an elevated quarter-wave vertical (whip mounted on a mast above ground), only 2–4 elevated radials are needed to achieve full efficiency. The reason is geometric: elevated radials operate as a counterpoise, not as a ground replacement. They establish the antenna’s λ/4-against-radials balanced relationship without interacting with lossy soil.
● top of vertical λ/4 element
│
│ vertical element (λ/4)
│
● feedpoint
╱ │ ╲
╱ │ ╲
╱ │ ╲ elevated radials (4×, each λ/4)
╱ │ ╲ sloped 30-45° downward
╱ │ ╲
╱ │ ╲
─────────────────────────── ground (~1 m below feedpoint)
radial tips do NOT touch ground
Elevated-radial efficiency vs # radials:
| # elevated radials | Efficiency | Notes |
|---|---|---|
| 1 | 30–50% | One radial = highly asymmetric; not recommended |
| 2 (opposite each other) | 75–85% | The minimum sensible configuration |
| 3 | 85–93% | Adds rotational symmetry |
| 4 | 90–95% | The standard — sweet spot |
| 6 | 93–96% | Marginal improvement |
| 8 | 94–97% | Diminishing returns |
The 4-radial elevated ground-plane outperforms a 16-radial buried-radial field in most installations. This is counterintuitive (more radials should be better, right?) but the elevated radials operate without ground-loss interaction. The buried field has to overcome typical-soil losses; the elevated field doesn’t.
The sloped-radial angle controls the feedpoint impedance:
- Horizontal radials (0° slope): feedpoint Z ≈ 25–32 Ω (mismatch on 50 Ω)
- Sloped 30° downward: Z ≈ 40–48 Ω (near match)
- Sloped 45° downward: Z ≈ 50–55 Ω (perfect 50 Ω match)
- Sloped 60° downward: Z ≈ 55–65 Ω (slight over-match)
A 45° downward slope gives a natural 50 Ω match with no transformer — the canonical “no-matching-network needed” elevated ground-plane configuration.
6.3 When to choose which
| Choose buried radials when | Choose elevated radials when |
|---|---|
| You can ground-mount the vertical | You need to mount on a roof, tower, or above obstructions |
| You have the space for radials extending outward | Lot constrains horizontal radial space |
| You’re willing to lay wire on/in soil | Soil is rocky, paved, or otherwise inaccessible |
| Local soil is moderate-to-good conductivity | Local soil is dry/rocky (elevated avoids ground loss entirely) |
| Multi-band operation (radials work on all bands equally) | Single-band or trap-multiband (elevated radials are band-specific) |
For multi-band trap verticals (Hustler 6-BTV, Cushcraft R-9), buried radials are the conventional choice because elevated radials would have to be sized differently for each band. Some commercial multi-band verticals (Cushcraft R-9) include a “no radial” matching network that approximates a counterpoise internally — this works but with a 1–3 dB efficiency penalty compared to a proper radial system.
Full radial-field treatment in Vol 20 §3.
7. Full-size HF trap verticals — Hustler 4-BTV / 5-BTV / 6-BTV / Cushcraft R-9 / GAP Titan
The mid-priced multi-band amateur HF vertical market is dominated by a handful of products, each with a distinct design philosophy. Knowing which compromises each makes helps pick the right one for your installation.
7.1 Hustler 4-BTV / 5-BTV / 6-BTV (the trap-vertical reference)
The Hustler BTV series uses series LC traps to give a single conductor multiple resonances — same principle as the trap dipole in Vol 7 §5. The 4-BTV covers 10/15/20/40 m (4 bands); the 5-BTV adds 30 m; the 6-BTV adds 17 m. None cover 80 m without the Hustler RM-80 add-on resonator coil (an extra $90).
| Model | Bands | Height | Power | Price (mid-2026) |
|---|---|---|---|---|
| Hustler 4-BTV | 10/15/20/40 m | 5.8 m (19 ft) | 1.5 kW | $280 |
| Hustler 5-BTV | 10/15/20/30/40 m | 7.2 m (24 ft) | 1.5 kW | $360 |
| Hustler 6-BTV | 10/12/15/17/20/30/40 m | 7.3 m (24 ft) | 1.5 kW | $430 |
| Hustler RM-80 (add-on) | adds 80 m | adds 1.5 m | 600 W | $90 |
Strengths: simple, classic, traps work reliably, mounting is straightforward (standard 1-1/2″ pipe at base), excellent customer support. The 6-BTV is one of the most-installed HF verticals in the world.
Weaknesses: per-band trap losses (~0.5–1 dB per band), narrow per-band bandwidth (~3%), requires a competent radial field, no 80 m without the add-on.
7.2 Cushcraft R-9 (the “no radials” premium vertical)
The Cushcraft R-9 (and its predecessor R-8) is an upper-tier vertical with a built-in matching network that partially compensates for the absence of a radial field. The R-9 covers 80–6 m on a 7.6 m (25 ft) vertical with internal traps and a base-mounted matching network.
Strengths: works (somewhat) without buried radials — appropriate for roof mounts and small lots; covers 80 m natively (unlike Hustler); 1.5 kW rating; well-built.
Weaknesses: $850+ price (more than 2× a 6-BTV); “no radials” claim is half-true — the antenna works, but with 2–4 dB efficiency loss compared to a properly grounded vertical. The internal matching network introduces its own losses. Several reviews note that adding 4 elevated radials to an R-9 noticeably improves performance — suggesting the “no radials” engineering is a compromise, not a miracle.
7.3 GAP Titan DX (the “no traps” vertical)
The GAP Titan DX takes a fundamentally different approach: instead of traps, it uses multiple parallel resonators — four separate radiating elements at four different lengths, sharing a common feedpoint. Each is resonant on its own band; the others present high impedance and stay out of the way (same principle as the fan dipole in Vol 7 §4, applied to a vertical).
Specs: covers 80/40/20/15/10 m, 7.6 m (25 ft) tall, no traps, no matching network (single direct-feed), 1.5 kW rating, $580.
Strengths: trap-free design eliminates the per-band trap-loss penalty; per-band efficiency is higher than the Hustler BTVs; 80 m is included natively. No matching-network losses.
Weaknesses: more wind-loaded than a single-tube vertical (multiple resonator wires); harder to install (more parts, more tuning); narrower per-band bandwidth than a Hustler (the parallel resonators interact a bit).
7.4 Cushcraft R-7000 (and predecessor R-7)
Predates the R-9; 40/30/20/17/15/12/10 m on a 7.0 m (23 ft) element; uses a different matching philosophy (“trapless” multi-section element with capacitive top loading). Discontinued but still common on the used market.
7.5 DX Engineering DX-MBVE-5
DX Engineering’s mid-tier vertical: 80/40/30/20/17/15/10 m on 8.2 m (27 ft), uses a base-mounted matching network and a single-element design with multiple resonant sections. $750. Excellent build quality; the “premium answer” without crossing into R-9 territory.
7.6 Decision matrix for mid-priced multi-band verticals
| Constraint | Best choice |
|---|---|
| Tight budget, 4–5 bands needed | Hustler 4-BTV ($280) |
| Moderate budget, all HF + WARC bands | Hustler 6-BTV ($430) |
| 80 m needed, willing to pay for quality | GAP Titan DX ($580) |
| 80–6 m on a small lot or roof, willing to pay premium | Cushcraft R-9 ($850) |
| Highest-quality build, willing to pay top dollar | DX Engineering DX-MBVE-5 ($750) or Cushcraft R-9 ($850) |
For a typical residential installation with the space for a 16–32 radial field: the Hustler 6-BTV is the price/performance sweet spot. For installations where buried radials are impossible (rooftop, paved yard, HOA-restricted): the Cushcraft R-9 is the proper compromise. For the operator who specifically needs 80 m DX without traps: the GAP Titan DX.
8. Shortened loaded verticals — coil/cap-hat loading and the efficiency penalty
When a full λ/4 vertical doesn’t fit, the answer is loading — adding a coil (inductor) or capacitive top-hat to electrically lengthen a shorter physical conductor into resonance. The compromise is efficiency: a shortened-loaded vertical radiates less than a full-size one, and the penalty grows rapidly as the antenna gets shorter relative to a quarter wave.
8.1 The shortened-vertical loading methods
There are four standard loading approaches, ranked by their efficiency:
Top-loaded (capacitive top hat):
- A horizontal “hat” of wires or a small umbrella structure at the top of the vertical adds capacitance to the high-voltage point.
- The capacitance effectively shifts the current maximum upward, lengthening the electrical antenna without lengthening the physical conductor.
- Highest efficiency of the loading methods: ~50–70% efficiency at λ/8 length.
- Common in commercial broadcast and amateur low-band verticals.
Center-loaded (inductor in the middle):
- A loading coil inserted halfway up the antenna.
- The coil’s reactance compensates for the missing λ/4 length, but the coil itself dissipates some of the current.
- ~15–30% efficiency at λ/8 length.
Bottom-loaded (inductor at the feedpoint):
- The simplest construction: a coil at the base in series with the antenna.
- Easiest to build and tune, but the coil sits at the current maximum (the feedpoint), so its losses are maximum.
- ~5–15% efficiency at λ/8 length. The worst of the four.
Top-hat + center coil (combined):
- A capacitive top-hat plus a smaller coil in the middle, hybrid design.
- 30–50% efficiency at λ/8 length.
- Common in commercial “compact verticals” like the MFJ-1796.
The efficiency table at λ/8 length:
| Loading method | Efficiency | Notes |
|---|---|---|
| Top-hat only (no coil) | 50–70% | Best — Marconi T antenna, “umbrella” antennas |
| Top-hat + small center coil | 30–50% | Hybrid — most commercial compact verticals |
| Center-loaded | 15–30% | Mid-range |
| Bottom-loaded | 5–15% | Worst — also the simplest to build |
The 80 m ham-stick (Vol 9 §4) is the textbook bottom-loaded example: 6 ft of physical length plus a base coil to electrically reach 130 ft (λ/4 at 3.65 MHz). The result radiates, but the efficiency is single-digit percent.
8.2 The fundamental efficiency limit
The Chu-Harrington limit (Vol 4 §6) sets a hard physics-of-small-antennas constraint: a shortened antenna’s bandwidth × efficiency is bounded by its physical volume. There is no clever-engineering escape — a smaller antenna must trade either bandwidth or efficiency.
For a shortened vertical the typical curve:
| Physical length (fraction of λ/4) | Best-case efficiency (top-hat loaded) | Bandwidth |
|---|---|---|
| 1.00 (full λ/4) | 95%+ | 5–8% |
| 0.75 | 85–90% | 4–6% |
| 0.50 | 75–85% | 3–5% |
| 0.33 | 60–75% | 2–4% |
| 0.25 (1/8 λ) | 50–70% | 1.5–3% |
| 0.10 | 25–40% | 0.5–1% |
| 0.05 | 5–15% | <0.5% |
The 80 m mobile ham-stick at ~6 ft physical length on a 130-ft full-size requirement is at 0.046 λ/4 — close to the bottom of this table. Single-digit-percent efficiency is what physics allows; there is no way to do better with a 6-ft mobile antenna on 80 m. The choice is “operate inefficiently or don’t operate at 80 m mobile.”
8.3 Top-hat construction
Top-hat geometry: 2–8 wires extending horizontally from the top of the vertical, each 1–3 m long. The wires don’t need to form a complete disk — sparse “spokes” work fine because the capacitance is bulk dielectric loading, not low-loss conductor area.
The classic amateur shortened vertical, the Marconi T, is a top-hat-loaded design: a vertical conductor with a horizontal top section that doubles as a transmission-line capacitor. The geometry is 1.5λ/4 of horizontal wire feeding into the top of a 0.25–0.5λ/4 vertical section, giving a low-impedance feed with respectable efficiency. The Marconi T is the standard low-band (160/80 m) home-station antenna at restricted-sized lots.
9. The half-square — a 1/4-wave vertical with 6 dB more gain
The half-square is one of the great underrated HF antennas — a simple geometry that turns two quarter-wave vertical elements connected by a half-wavelength horizontal top wire into an antenna with 5 dB more gain than a single vertical and 6 dB of front-to-back ratio. Designed by Harry “Ben” Black W4RNL (extending earlier work by William Orr W6SAI) in the late 1970s.
9.1 Geometry
Two λ/4 vertical wires connected at the top by a λ/2 horizontal wire, fed at one of the bottom corners:
●━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━● ← horizontal λ/2 wire (top)
│ │
│ │
│ λ/4 vertical │ λ/4 vertical
│ │
●━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ground
↑
feedpoint (one corner)
The currents in the two vertical elements are in-phase (driven by the same feedline through the top connection), giving constructive addition broadside to the plane of the antenna. The pattern is a directional figure-8 with peak gain ~5 dBd (≈ 7.15 dBi) and 6 dB of front-to-back ratio.
9.2 What you get
- +5 dBd gain over a single vertical (so ~5 dBi gain over real ground)
- Front-to-back ratio: 6 dB
- Vertical polarization — same as a single vertical, useful for DX paths
- Lower feedpoint impedance: ~25–35 Ω at the corner feed, needs a 1.5:1 transformer or shunt-L match
- No radials required at the bottom of the vertical elements — the half-square is self-contained, similar to a vertical dipole
- Wider bandwidth: ~10% (better than a single trap vertical)
9.3 Construction and deployment
The half-square is bigger than a quarter-wave vertical: needs two support points (one for each top corner of the horizontal wire) at least λ/4 above ground, with a λ/2 horizontal separation between them. For 20 m, the geometry is:
- Two vertical wires, each ~5 m long
- Horizontal connecting wire, ~10 m long
- Total volume occupied: 10 m × 0.5 m × 5 m
- Needs two supports at ~5 m height, 10 m apart
For 40 m, scale up: 10 m verticals, 20 m horizontal, ~10 m supports. For 80 m, scale to 20 m verticals, 40 m horizontal, ~20 m supports — at this size the half-square is a serious installation but still smaller than a full-size 80 m vertical with a radial field.
The half-square’s main niche: a half-square covers the same ground as a 2-element Yagi, at a fraction of the cost and complexity, with the directional advantage that a single vertical doesn’t have. For a moderate-size lot with two trees 10 m apart, a 20 m half-square is a $50 build that outperforms most commercial verticals.
10. Best-case use
- HF DX from a small lot with no horizontal real estate but enough vertical clearance. A 40 m vertical on a 10 ft mast outperforms a 40 m dipole at 30 ft for low-angle DX.
- Multi-band operation with a single feedline: a Hustler 6-BTV or GAP Titan DX is the canonical multi-band vertical for an operator who wants HF coverage without per-band antenna switching.
- 80 m / 160 m where horizontal dipoles are impractical due to height — a shortened-loaded 160 m vertical (e.g. an Inverted L: vertical + horizontal at the top) is the standard low-band antenna for residential lots.
- Beach or waterfront operation: salt-water ground is the rare case where a vertical’s low-angle pattern reaches DX-grade gain (5+ dBi). A vertical at the beach is a different antenna from a vertical in the suburbs.
- DXCC-chasing on a small budget: vertical + 32 radials beats most $400–600 antennas for low-angle DX work. Total parts cost ~$120 (vertical + wire + connectors).
- Roof-mounted or tower-mounted multi-band antennas: the elevated ground-plane is the textbook roof-vertical solution — 4 elevated radials, no buried radial field, clean low-angle pattern.
- Field portable serious-DX: a knock-down vertical with elevated radials assembled in 20 minutes is a competition-grade field antenna. The Vibroplex Field-vertical and SOTAbeams SOTApole are the canonical references.
11. Worst-case use
- Stealth deployments: a 25 ft (8 m) vertical visible from blocks away. HOA-restricted lots can’t accommodate it; an indoor dipole or a stealth random-wire (Vol 10) is the workaround.
- Dry, sandy, rocky soil without compensation: the radial system can’t fully replace ground in pathological-conductivity soils. A vertical in Nevada desert without a buried radial field is a heating element more than a radiator.
- High local noise environments: verticals receive vertical-polarization noise preferentially; if you live near a switching power supply, your noise floor is 10–15 dB worse on a vertical than on a horizontal dipole.
- Casual HF, single-band, “first antenna” use: a half-wave dipole at 30 ft is simpler, cheaper, more forgiving, and better for the operator’s first antenna. Verticals reward attention to the radial system; dipoles are more idiot-proof.
- Short-haul regional operation: NVIS works on 40 m and below with a low horizontal dipole, not a vertical. A vertical’s pattern null points straight up — the exact NVIS-required direction.
- Mountain-top portable HF: a vertical’s radial field is too cumbersome to deploy on a summit; SOTA/POTA activations almost always use a portable dipole (Vol 7 §8 linked dipole) instead.
12. Power handling
A vertical’s power handling lives in three places:
| Limit | Typical max continuous | Notes |
|---|---|---|
| Wire/element conductor heating | 1.5–2 kW for 1/2″ tubing; 800 W for #14 wire | Conductor diameter dictates the limit |
| Trap dissipation (trap verticals) | 800 W–1.5 kW depending on number of traps active | Traps fail open when overloaded (audible “ping” then SWR to infinity on that band) |
| Base insulator / matching network | 1.5 kW typical | Lossy matching network can over-dissipate on off-resonance bands |
Specific commercial-vertical ratings (mid-2026):
| Vertical | Power rating | Limit |
|---|---|---|
| Hustler 6-BTV | 1.5 kW SSB | Trap construction |
| Cushcraft R-9 | 1.5 kW SSB | Matching network |
| GAP Titan DX | 1.5 kW SSB | Wire/feedpoint |
| Cushcraft R-7000 (older) | 1 kW SSB | Older trap design |
| DX Engineering DX-MBVE-5 | 1.5 kW SSB | Conservative |
| DIY 20 m elevated GP (1/2″ tubing) | 2 kW SSB | Conductor-limited |
| 80 m shortened ham-stick | 200 W | Coil saturation |
For amateur amplifier (1.5 kW) operation, the GAP Titan DX or DX Engineering DX-MBVE-5 are the conservative choices; the Hustler 6-BTV at full power runs the traps at their limit and shortens service life (traps fail open after enough cycles at full power).
13. DIY build — a 20 m elevated quarter-wave ground-plane
This is a serious-DX antenna at a casual-amateur price. About 4 hours of work plus tuning time. Total parts cost ~$130 USD.
13.1 Geometry and parts
A 20 m vertical with 4 elevated radials sloped at 45° downward from horizontal — the canonical no-buried-radials configuration. Feedpoint is naturally close to 50 Ω.
| Part | Specification | Source | Mid-2026 price |
|---|---|---|---|
| Vertical element | 1/2″ aluminum tubing, 5.0 m total, telescoping | Texas Towers TT12HD aluminum mast section, or DX Engineering DXE-WHIPV-5 | $45–60 |
| Radial wires (4×) | #14 AWG copper, ~5.0 m each (20 m total) | Wireman 534 hard-drawn ($0.55/ft) | $11 |
| Feedpoint base | Insulated SO-239 + radial-attachment ring + 1.5:1 unun | DX Engineering DXE-MFTC1.5 ($75) or DIY (Hammond box + parts, $25) | $25–75 |
| Mast or support | Telescoping aluminum pole (extends to ~1.5 m) | Spiderbeam mini-mast, or DIY from 1″ aluminum pipe | $30 |
| Coax pigtail | RG-8X, 1 m, with PL-259 | Times Microwave | $15 |
| Hardware | Stainless U-bolts, hose clamps, terminal lugs | Various | $10 |
| Weatherproofing | Coax-Seal + 3M 33+ | $10 | |
| Total (commercial DXE base) | ~$176 | ||
| Total (DIY base) | ~$126 |
13.2 Step-by-step construction
Cut the vertical element. A 20 m quarter-wave is 5.03 m for k=0.95 (1/2″ tubing). Cut to 5.05 m (about 0.4% long for trim headroom). Telescoping aluminum lets you fine-tune by adjusting the upper section’s overlap.
Cut the radial wires. Each is 5.03 m (16.5 ft) of #14 AWG copper. Cut all four to identical length within ±5 mm — radial-length asymmetry distorts the pattern.
Build the feedpoint. The feedpoint is an SO-239 mounted in an insulated mounting bracket, with:
- Center conductor connected to the bottom of the vertical element (via a flexible braided strap so the vertical can be removed for maintenance)
- Shield connected to a common bus that’s bonded to the four radial-attachment points
For the commercial DXE-MFTC1.5, the unun and feedpoint are pre-built; mount the vertical and connect radials. For DIY, mount the SO-239 in a polycarbonate Hammond box (1591-N or similar), with feedthrough bushings for the radial connections.
Mount the support. Bolt the feedpoint base to the top of the telescoping support mast (~1.5 m above ground). Extend the mast so the vertical’s base sits at ~2 m above ground; the radials will slope downward from there.
Install the radials. Bolt each radial wire to its radial-attachment point at the feedpoint base. Run each radial outward and downward, sloping at 45° from horizontal — the tip of each radial should be about 2.5 m horizontal distance from the base and 1 m below the feedpoint (still ~1 m above ground). Use guy-line stakes at the radial tips to keep them under tension.
Hoist the vertical. Slide the vertical element into its mounting bracket; secure with the U-bolt. Final height: ~7 m total (1.5 m support + 5 m vertical).
Sweep and trim. Connect the NanoVNA at the feedline end. Sweep 13.5–14.7 MHz. Look for SWR minimum near 14.175 MHz. If resonance is too low (e.g. 13.9 MHz), the vertical is too long — trim 2–3 cm at the top section’s upper joint. If too high (e.g. 14.4 MHz), extend the upper section. The trim sensitivity at 14 MHz is ~3 cm per 50 kHz of shift.
Lock and weatherproof. Once tuned, apply 3M Scotch 130C tape to all outdoor connections. Final SWR target: <1.3:1 at 14.175 MHz, with the 2:1 bandwidth covering ~400 kHz of the band (~13.95 to 14.35 MHz).
13.3 Tuning verification
A successful elevated-ground-plane build shows:
- SWR minimum < 1.3:1 at the target frequency
- Resistive feedpoint impedance (X ≈ 0 on the Smith chart at minimum)
- R ≈ 45–55 Ω at minimum — naturally close to 50 Ω because of the 45° radial slope
- 2:1 SWR bandwidth ~400 kHz around the design frequency
- No anomalous resonances between half-wave and full-wave (e.g. at ~21 MHz or ~28 MHz — small spikes there indicate parasitic resonances from the radial geometry; usually harmless)
If the SWR minimum is far off the expected R value (e.g. R = 15 Ω with very low SWR), the matching is incorrect — check the radial slope angle and the 1.5:1 unun’s connection direction.
14. Commercial buys
Sorted by tier and use case (USD, mid-2026):
| Tier | Model | Bands | Price | Notes |
|---|---|---|---|---|
| Budget | MFJ-1796 | 6/10/15/20/40 m (compact) | $230 | Compact (4 m tall) shortened-loaded vertical; modest efficiency on 40 m |
| Budget | Diamond CP-510M | 6 m + 50 MHz (mono) | $180 | High-quality VHF/6m vertical |
| Budget | HyEndFed verticals (Trans-World) | various single-band | $200–350 | EU manufacturer, excellent build |
| Mid | Hustler 4-BTV | 10/15/20/40 m | $280 | Classic trap vertical reference |
| Mid | Hustler 6-BTV | 10/12/15/17/20/30/40 m | $430 | Most-installed amateur HF vertical |
| Mid | GAP Titan DX | 80/40/20/15/10 m | $580 | Trap-free; the “engineering-quality” mid-tier choice |
| Mid | Cushcraft R-9 | 80–6 m | $850 | Premium “no radials” design |
| Premium | DX Engineering DX-LB-MS | 80/40/30/20/17/15/10 m | $1100 | DX Engineering’s house design; serious build |
| Premium | DX Engineering DX-MBVE-5 | 80–10 m | $750 | Mid-tier DX Engineering |
| Premium | Hy-Gain AV-680 | 8-band, no traps | $1200 | Multi-section vertical; commercial-grade |
| Premium | SteppIR BigIR | 40–6 m (mechanically resonant) | $1850 | Motor-driven tape-measure-element vertical; effective bandwidth = motor-controlled |
| Premium | Mosley TA-32 | 20/15/10 m (3-band, no traps) | $1400 | Heavy-duty, multi-element design |
What to avoid:
- “Multi-band miracle verticals” priced under $150 — usually they’re shortened-loaded with mediocre efficiency. The economics don’t allow a full-size multi-band vertical under $200.
- Cheap “no radial” verticals without published matching-network spec — most achieve “no radials” by being lossy enough that the lossy matching network sets the feedpoint Z near 50 Ω regardless of what the antenna does. The SWR looks fine; the efficiency is awful.
- “Compact 80m verticals” shorter than 8 m — physics says these are < 30% efficient on 80 m. They work but they’re not what they claim.
Used market is excellent for verticals: Hustler 4-BTVs and 6-BTVs from the 1980s–2000s still work fine after 30 years if traps haven’t failed. A used 6-BTV at $150 is the best HF-vertical bargain available.
15. Companion gear — radial field, mounting, feedline
A vertical antenna is a system; the visible whip is half of it. The other half is the radial field plus the supporting infrastructure:
- Radial wire: #14 or #16 copper-clad steel (CCS) — the broadcast-industry standard, cheap ($0.20/ft), corrosion-resistant under burial. DX Engineering DXE-RADIAL-1, Wireman 522 (#16 CCS).
- Radial bus/common point: a corrosion-resistant brass or stainless terminal block bonded to the coax shield and to all radials. DX Engineering DXE-RAD-CON, MFJ-1929, or DIY from a brass block with stainless screws.
- Mounting: ground-mounted verticals need a non-conductive insulator at the base (so the radial system, not the vertical itself, is the RF ground reference). Common: 1″–2″ PVC pipe coupler + a stainless threaded rod into the ground.
- Feedline: RG-8X for low-power runs <25 m; LMR-400 for permanent or higher-power runs; hardline (LDF4-50A) for the obsessive. The feedline-loss budget is in Vol 5 §6.
- Common-mode choke: a 1:1 current choke (Mix-31 toroid or string-of-beads) at the feedpoint suppresses common-mode currents on the coax shield. More important for elevated verticals than for ground-mounted ones.
- Lightning protection: a polyphaser arrestor at the feedline’s shack entry, grounded to a single-point ground. Verticals are more lightning-prone than dipoles (they’re vertical conductors against ground — exactly the geometry a strike likes) — don’t skip this. Vol 20 §5 covers single-point grounding in depth.
- Mast or support: a non-conductive (fiberglass) or insulator-isolated support to mount the vertical at the right height. For elevated GPs, a 1–2 m support raises the feedpoint above grass and small obstacles.
- Guying or stay system: a 40 m vertical (5+ m tall) catches wind. A simple Dacron-rope guy system at the top reduces wind-loading deflection and improves long-term mechanical stability.
16. Common gotchas and myths
-
“No radials needed” — a marketing lie or a half-truth. Some verticals work without buried radials (Cushcraft R-9, GAP Titan with limited efficiency), but no vertical works as well without a radial system as it does with one. The “no radials” claim usually means “the antenna has internal lossy matching that compensates by burning power instead of radiating it.”
-
“Verticals are inferior to horizontal antennas” — only in receive (where they pick up more vertical-polarization noise). In transmit for DX, a vertical over a competent ground system matches or beats a low horizontal dipole at low elevation angles. The “verticals are worse” belief comes from operators who’ve used verticals without proper radial systems and compared them to dipoles that didn’t need similar tuning.
-
“More radials = always better” — diminishing returns after ~64 buried radials. The 65th radial adds <0.3 dB to your signal. The 120th adds <0.1 dB. Save the labor for other improvements.
-
“My SWR is 1.5:1 so it’s working” — SWR alone doesn’t measure efficiency. A vertical with 4 buried radials over poor soil shows 1.5:1 SWR because ground loss adds resistance; the antenna works but it’s only 30% efficient. A vertical with 60 radials shows 1.2:1 SWR but is 80% efficient. The lower SWR is worse for power transfer.
-
“The Hustler 6-BTV is a 7-band antenna” — it’s a 6-band antenna for the four high bands (10/15/20/40) where the traps work well, and three more bands at compromised performance. The 80 m option requires the RM-80 add-on (not 80 m natively).
-
“Vertical = omnidirectional” — only over a clean uniform ground. Real installations near buildings, towers, or trees produce azimuth-pattern distortion of 2–5 dB. The “omnidirectional” pattern is a theoretical idealization.
-
“Quarter-wave verticals have 5.15 dBi gain” — over a perfect ground (theoretical infinite-conductivity image plane). Over typical real ground with a competent radial system, peak gain is 2–4 dBi. The 5.15 dBi figure is rarely seen outside salt-water-beach installations.
-
“Verticals are stealthy” — no. A 25-ft vertical visible from blocks away is the opposite of stealthy. Stealth installations use random-wire (Vol 10) or attic dipoles, not visible verticals.
-
“The radial field doesn’t need to be tuned” — true for buried radials (the soil dampens any resonance). False for elevated radials: each elevated radial should be cut to λ/4 at the design frequency. Asymmetric elevated radials distort the pattern.
-
“I can convert a buried-radial vertical to elevated by lifting the radials” — not quite. Buried-radial verticals are typically tuned for the buried-radial feedpoint impedance (slightly different from elevated). Lifting the radials shifts the feedpoint Z; you’ll need to re-trim or re-match.
17. Resources
- ARRL Antenna Book Ch. 5 (verticals) — the canonical multi-band-vertical reference.
- Sevick, Building and Using Baluns and Ununs — radial-field optimization deep dive.
- ON4UN, Low-Band DXing (5th ed.) — the authoritative 80m/160m vertical reference for serious DXers.
- L. B. Cebik (W4RNL) papers on vertical antenna modeling — antenneX library.
- Brown / Lewis / Epstein 1937 paper on broadcast antenna radial fields — the original “120 radial” recommendation paper; still cited.
- DX Engineering Tech Articles — modern vertical installation notes.
- Cushcraft and Hustler manuals — manufacturer-published tuning and installation procedures.
- W6SAI Yagi book (older but classic) — has substantial vertical-antenna coverage.
- Maxwell’s Reflections III — clarifies the SWR/efficiency relationship that the vertical-vs-radial-system question keeps confusing.